File size: 1,117 Bytes
7d8d7c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
---
license: apache-2.0
base_model: Wan-AI/Wan2.1-T2V-1.3B
pipeline_tag: video-to-video
library_name: pytorch
tags:
- video-generation
- camera-control
- wan
- rerope
---

# ReRoPE V2V checkpoint

Inference-only V2V weights for **ReRoPE**, based on
[Wan2.1-T2V-1.3B](https://huggingface.co/Wan-AI/Wan2.1-T2V-1.3B).

## Files

| File | Parameters | Dtype | Size | SHA-256 |
| --- | ---: | --- | ---: | --- |
| `rerope_v2v_wan21_1.3b.safetensors` | 1,418,996,800 | bfloat16 | 2,838,077,456 bytes | `31f74e89840b9ff213a4adf22c0e31d69ec5e6edacf793b00c51c475767d6ee8` |

The file contains 825 tensors and only the inference state dictionary. Optimizer
state, trainer state, datasets, and local paths are not included.

## Usage

Use these weights with the inference code at
[github.com/Sisyphe-lee/ReRoPE_](https://github.com/Sisyphe-lee/ReRoPE_).

The Wan2.1 base model is required separately. See the GitHub README for the
expected directory layout and V2V command.

## Scope

This initial model release contains one V2V checkpoint. The repository includes
I2V inference code, but no I2V ReRoPE checkpoint is released.